Research map / Detection and segmentation / CNN features
Teaching AI the Anatomy Behind the Scan: Addressing Anatomical Flaws in Medical Image Segmentation with Learnable Prior
Young Seok Jeon, Hongfei Yang, Huazhu Fu, Mengling Feng
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add the Deform block to the local prior (ablation) | AIC-Net with UNet · TotalSegmentator abdominal organs (local prior accuracy) | DSC: 60.8 -> 67.6 | Improved |
| add the Deform block to the local prior (ablation) | AIC-Net with DeepLabV3+ · TotalSegmentator abdominal organs (local prior accuracy) | DSC: 57.5 -> 68.1 | Improved |
| add the Deform block to the local prior (ablation) | AIC-Net with UNETR · TotalSegmentator abdominal organs (local prior accuracy) | DSC: 57.2 -> 65.9 | Improved |
| add the Deform block to the local prior (ablation) | AIC-Net with UNETR-Swin · TotalSegmentator abdominal organs (local prior accuracy) | DSC: 60.6 -> 70.7 | Improved |
| add the Deform block to the local prior (ablation) | AIC-Net with UNet · TotalSegmentator vertebrae (local prior accuracy) | DSC: 55.4 -> 68.9 | Improved |
| add the Deform block to the local prior (ablation) | AIC-Net with DeepLabV3+ · TotalSegmentator vertebrae (local prior accuracy) | DSC: 53.4 -> 64.2 | Improved |
| add the Deform block to the local prior (ablation) | AIC-Net with UNETR · TotalSegmentator vertebrae (local prior accuracy) | DSC: 45.5 -> 58.1 | Improved |
| add the Deform block to the local prior (ablation) | AIC-Net with UNETR-Swin · TotalSegmentator vertebrae (local prior accuracy) | DSC: 51.2 -> 71.0 | Improved |
| add the learnable prior and Deform block to the cascaded model (UNETR) (ablation) | UNETR cascaded · TotalSegmentator vertebrae | DSC: 76.2 -> 83.2 | Improved |
| add the learnable prior and Deform block to the cascaded model (DeepLabV3+) (ablation) | DeepLabV3+ cascaded · TotalSegmentator abdominal organs | HD95: 7.28 -> 4.28 | Improved |
| add the centroid loss (ablation) | AIC-Net prior learning · TotalSegmentator vertebrae | learned prior quality (qualitative) | Improved |
| add the learnable prior to a plain backbone with global-local cascade (UNETR-Swin) (ablation) | UNETR-Swin · TotalSegmentator vertebrae | HD95: 12.6 -> 1.76 | Improved |
Closest papers by meaning
- Anatomical Priors in Convolutional Networks for Unsupervised Biomedical Segmentation (CVPR 2018 · similarity 0.89)
- Prior-Aware Neural Network for Partially-Supervised Multi-Organ Segmentation (ICCV 2019 · similarity 0.89)
- Mine yOur owN Anatomy: Revisiting Medical Image Segmentation With Extremely Limited Labels (TPAMI 2024 · similarity 0.88)
- Continual Segment: Towards a Single, Unified and Non-forgetting Continual Segmentation Model of 143 Whole-body Organs in CT Scans (ICCV 2023 · similarity 0.88)
- CAT: Coordinating Anatomical-Textual Prompts for Multi-Organ and Tumor Segmentation (NeurIPS 2024 · similarity 0.86)
- Upping the Game: How 2D U-Net Skip Connections Flip 3D Segmentation (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
